The aim of this work is to produce monoclonal antibody-based agents for use in early detection of cancer by positron emission tomography (PET). The specific agent we will investigate in this phase I work will be a complementarity-determining region-grafted antibody targeted against carcinoembryonic antigen radiolabeled with zirconium. The preferred bifunctional chelate for zirconium radiolabeling of proteins will be found. Radiolabeling, stability, in vitro binding and in vivo disposition of gamma-ray-emitting zirconium-88-radiolabeled antibody IgG, F(ab')2 and Fab' moieties will be studied to determine the probable optimal antibody form. In parallel, methodologies to produce and process the positron- emitting zirconium-89 nuclide will be developed. At the end of the phase I period the zirconium-89 will be coupled to the optimal antibody form to demonstrate the ability of the agent to successfully image tumor xenografts in animals. With feasibility demonstrated, in SBIR phase II work the method will be extended to other humanized monoclonal antibodies. At least one of these will then be entered into phase I clinical trials to establish human imaging utility and, if that is proven, the agent will be developed on a commercial production scale. Ultimately, non-immunogenic specific targeting agents for PET imaging of many cancers are envisioned. PROPOSED COMMERCIAL APPLICATIONS: PET is considered to be a most sensitive imaging modality. Monoclonal antibodies offer targeting specificity to defined diseases and when humanized, are essentially non-immunogenic. By coupling an appropriate PET radionuclide with humanized monoclonal antibodies directed against diverse cancers, highly, specific and sensitive imaging agents for the early detection of new and recurrent cancers would become available. Very significant clinical impact and great commercial potential exists for such agents.